A significant proportion of patients treated with Androgen deprivation (AD) for prostate cancer ultimately develop progressive disease in the setting of anorchid testosterone levels. 2 mechanisms that have been proposed for the failure of AD are 1) stimulation of the androgen receptor (AR) by the adrenal androgens androstenedione and dehydroepiandrosterone (DHEA) and 2) stimulation of the AR by growth factors such as the insulin like growth factor- type 1 (IGF-1) and its receptor, IGF-1R. Preliminary data suggests that the activity of the IGF axis is partially regulated by androgen. It is hypothesized that blockade of both adrenal androgens and the IGF-1R will improve clinical outcomes in patients. Studies in cell culture indicate that the small molecule meso-nordihydroguariacetic acid (NDGA) inhibits the IGF-1R in cancer cells. Since this agent appears safe in preclinical models, a phase I study in relapsed prostate cancer patients to determine the maximum tolerated dose (MTD) and PSA modulating effects after 28 days of treatment is planned. In parallel with this, a murine model will be developed in which the targeting of AR stimulation by both androgen and IGF-1 with NDGA can be explored. Mice bearing androgen responsive LAPC-9 human prostate cancer cells will be treated with combinations of AD, NDGA and the drug Pioglitazone (to reduce circulating free IGF ligand). Similarly, it is hypothesized that the IGF-R signaling can be attenuated in prostate by pharmacologically reducing circulating IGF levels in addition to receptor blockade with NDGA. Phase II studies of NDGA alone and combined with Pioglitazone will be performed in patients with androgen independent prostate cancer utilizing standard efficacy criteria. In parallel with these studies, AR signaling by the adrenal androgens will be targeted. A widely utilized approach to reducing adrenal androgen production is the use of ketoconazole, an inhibitor of adrenal steroid biosynthesis. Despite high initial responses, resistance to this therapy develops in most patients, limiting its long-term clinical effectiveness. It is hypothesized that resistance to ketoconazole occurs through adrenal hyperplasia, and that this resistance can be overcome by suppressing adrenocorticotripic hormone (ACTH) secretion with dexamethasone. In a phase II trial, the effects of ketoconazole on ACTH levels over time will be measured and the efficacy of combining ketoconazole with total ACTH suppression with dexamethasone at the time of clinical progression will be measured. Subsequently, based on the results of these early studies, attempts will be made at combination therapy with NDGA, Pioglitazone, Ketoconazole and dexamethasone in order to maximally target the interaction of androgen and signaling by the Insulin -like growth factor receptor.